Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare readOnly/openWorld/non-destructive, so the description adds real value beyond them: it discloses that results are historical database facts rather than live model output (a staleness caveat), and that filters behave differently by kind (domain filters hit cited sources, keyword filters hit answer text). The instruction to "preserve nullable evidence and unsupported dimensions" usefully warns that fields may be null or absent and should not be inferred or backfilled.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Is the description appropriately sized, front-loaded, and free of redundancy?

Four short sentences, front-loaded with what the records are and then the caveats. The telegraphic style ("Historical provider database facts, not a live model execution") is dense but each sentence carries distinct information. Minor awkwardness, no filler.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a read-only tool with no output schema and 0% schema coverage, the description covers data provenance and nullability well but omits the shape of returned records beyond a phrase, plus the platform dimension that materially changes results. An agent can call it, but cannot predict the response or the effect of platform/limit without experimenting.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

With 0% schema description coverage, the description compensates for the most important parameter by explaining what the DOMAIN vs KEYWORD target kinds match against — a meaningful semantic addition. However, it says nothing about limit (cap of 20), market/language (both pinned to US/en), or platform (google vs chat_gpt), so four of five parameters remain semantically dark.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Does the description clearly state what the tool does and how it differs from similar tools?

The first sentence is a noun phrase listing record contents ("cited sources, search references and brand entities") rather than a verb+resource statement, leaving the retrieval action implicit in the name. The resource is identifiable, but the description never distinguishes this from close siblings like nexscope_seo_ai_mention_history or nexscope_seo_ai_mention_deltas. Adequate but vague on the operation.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Does the description explain when to use this tool, when not to, or what alternatives exist?

"Historical provider database facts, not a live model execution" implies the when-to-use boundary (use this for snapshot data, use a live-observation tool otherwise), but no sibling is named and no exclusion is explicit. An agent must infer that nexscope_seo_model_answer_observation covers the live case. Implied usage only.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.